The objective of this proposal is to determine the feasibility of a new tool that enables augmentative communication users to communicate more efficiently. This improvement in efficiency is accomplished by designing a computer-based listening system that monitors the conversation between a speaking communication partner and a typing augmentative communication (AAC) user. The computer uses the information it "hears" from the speaking partner to provide the AAC user with a list of topic- specific responses. The AAC user utilizes this list to communicate more quickly with the speaking partner. This topic prediction approach may significantly reduce the number of keystrokes required of AAC users in order to express themselves. This increase in efficiency will enable the user to communicate more quickly, addressing the single greatest challenge facing the AAC field. PROPOSED COMMERCIAL APPLICATION: Computer-based augmentative communication systems continue to gain acceptance and market share. This research can be applied to most augmentative communication systems, providing the opportunity for implementation in several products. The commercial potential is excellent.